本論文提出兩種方法應用於卷積神經網路局部放電檢測系統。第一種方法為使用孤立森林算法，檢測離群值以改善局部放電相位解析圖(PRPD)正規化後，而使得局部放電信號波型過小的問題。第二種方法為利用DS理論(Dempster–Shafer theory)作為決策融合的規則，將多個分類器所得出的決策，組合在同一個特徵向量之中，得出最後共同的決策。本系統以高頻比流器(HFCT)、暫態對地電壓感測器(TEV)和超高頻感測器(UHF)作為感測器量測電力設備之局部放電訊號。數據會經由帶通濾波器(Band-pass Filter)與小波閥值法(Wavelet Threshold) 去除雜訊，以及透過提出的孤立森林算法檢測離群值，以蒐集到的相位、振幅資料做正規化並繪製局部放電相位解析圖做為特徵輸入到卷積神經網路進行辨識，並透過提出的DS理論分辨局部放電類型。

In this thesis, approaches for the partial discharge detection system of the convolutional neural network (CNN) are proposed. The first approach would be to use the isolation forest algorithm to identify outliers in order to address the issue of the partial discharge signal waveform being too small after normalizing the phase resolution partial discharge (PRPD). The second approach would be to apply Dempster–Shafer theory as the rule of decision fusion and integrate the decisions produced by various classifiers into the same eigenvector in order to arrive at the final joint decision. As sensors for measuring the partial discharge signal of power equipment, this system utilizes a high frequency current transformer (HFCT), transient earth voltage (TEV), and ultra-high frequency (UHF). The Band-pass Filter and Wavelet Threshold would remove the noise, and the Isolation Forest algorithm would identify outliers. The acquired phase and amplitude data would be normalized, and the phase resolution partial discharge (PRPD) would be utilized as a recognition feature input to the convolutional neural network (CNN). Using the DS theory, the type of partial discharge would be distinguished.

1. R. Bartnikas, J. P. Novak, “On the Character of Different Forms of Partial Discharge and their related Terminologies” IEEE Transactions on Electrical Insulation,vol. 28,no. 6,pp. 956-958,1993.
2. P. H. F. Morshuis, “Partial Discharge Mechanisms in Voids related to Dielectric Degradation” IEEE Proceedings – Science,Measurement and Technology, vol. 142, no.1 , pp. 62-68 ,1995.
3. R. Yao, J. Li, M. Hui, L. Bai, and Qisheng Wu, “Feature Selection Based on Random Forest for Partial Discharges Characteristic Set” IEEE Access, vol. 8, pp.159151 - 159161,2020.
4. M. Elsisi, M. Tran,K. Mahmoud, D. A. Mansour, M. Lehtonen , and M. M. F. Darwish, “Towards Secured Online Monitoring for Digitalized GIS Against Cyber-Attacks Based on IoT and Machine Learning ” IEEE Access, vol. 9, pp. 78415 - 78427,2021.
5. Y. LeCun, et al. “Backpropagation applied to Handwritten Zip Code Recognition”,Neural computation, vol. 1, no. 4, pp.541-551, 1989.
6. A.Gao, Y. Zhu, W. Cai , and Y. Zhang. “Pattern Recognition of Partial Discharge Based on VMD-CWD Spectrum and Optimized CNN With Cross-Layer Feature Fusion,” IEEE Access, vol. 8, pp. 151296-151306, 2020
7. F.-C. Gu, “Identification of Partial Discharge Defects in Gas-Insulated Switchgears by Using a Deep Learning Method,” IEEE Access, vol. 8, pp. 163894-163902, 2020.
8. B. An, S. Yoon, S. Lee, J. Kim, Y. Jung, and T. Jang. “Effect of Background Noise Discrimination on Partial Discharge Pattern Recognition using Neural Network and Support Vector Machine,” 2020 International Symposium on Electrical Insulating Materials, 2020.
9. X. Peng et al., “A Convolutional Neural Network-Based Deep Learning Methodology for Recognition of Partial Discharge Patterns from High-Voltage Cables,” IEEE Transactions on Power Delivery, vol. 34, pp. 1460-1469, 2019.
10. M. H. Wang, S. D. Lu and R. M. Liao. “Fault Diagnosis for Power Cables Based on Convolutional Neural Network with Chaotic System and Discrete Wavelet Transform,” IEEE Transactions on Power Delivery, vol.37. pp. 582 – 590,2021.
11. T.-D. Do, V.-N. Tuyet-Doan, Y.-S. Cho, J.-H. Sun, and Y.-H. Kim, “Convolutional-Neural-Network-Based Partial Discharge Diagnosis for Power Transformer Using UHF Sensor,” IEEE Access, vol. 8, pp. 207377-207388, 2020.
12. J. Li, X. Han, Z. Liu and X. Yao, “A Novel GIS Partial Discharge Detection Sensor with Integrated Optical and UHF Methods,” IEEE Transactions on Power Delivery, vol. 33, pp.2047-2049,2018.
13. 林鈺洋，「基於卷積神經網路與資料融合技術之局部放電檢測系統」國立成功大學電機工程學系碩士論文，2020年。
14. 黃明義，「不受相位影響之局部放電信號相位特徵用於卷積神經網路之絕緣瑕疵類型」國立成功大學電機工程學系碩士論文，2021年。
15. Liu, F. Tony, K. M. Ting, and Z.H. Zhou. “Isolation Forest,” Eighth IEEE International Conference, 2008.
16. G. Shafer. “A Mathematical Theory of Evidence, “Princeton University Press, ”1976.
17. F. Álvarez, et.al “Application of HFCT and UHF Sensors in On-Line Partial Discharge Measurements for Insulation Diagnosis of High Voltage Equipment,” Sensors, vol. 15, pp. 7360-7387,2015.
18. A. Eigner, K. Rethmeier, “An Overview on the Current Status of Partial Discharge Measurements on AC High Voltage Cable Accessories,”IEEE Electrical Insulation Magazine, vol. 32, pp. 48-55,2016.
19. D. A. Nattrass, “Partial Discharge Measurement and Interpretation,” IEEE Electrical Insulation Magazine, vol. 4, no.3, pp .10-23,1988.
20. W. Wang, et al, “Anomaly Detection of Industrial Control Systems Based on Transfer Learning” Tsinghua Science and Technology, vol. 26, pp.821 - 832,2021.
21. J. Hong, C.-C. Liu, and M. Govindarasu, “Integrated Anomaly Detection for Cyber Security of the Substations” IEEE Transactions on Smart Grid, vol.5, pp. 1643 - 1653, 2014.
22. A. R. Javed, et al, “Anomaly Detection in Automated Vehicles Using Multistage Attention-Based Convolutional Neural Network” IEEE Transactions on Intelligent Transportation System, vol.22, pp. 4291-4300, 2021.
23. D. L. Hall, J. Llinas, “Multisensor data fusion”, 2001.
24. M. A. Al-Jarrah, M. A. Yaseen, A. A. Dweik, O. A. Dobre, and E. Alsusa, “Decision Fusion for IoT-Based Wireless Sensor Networks” IEEE Internet of Things Journal, vol.7, pp.1313 -1326,2020.
25. W. Li, S. Prasad, J. E. Fowler, “Decision Fusion in Kernel-Induced Spaces for Hyperspectral Image Classification,” IEEE Transactions on Geoscience and Remote Sensing, vol. 52, pp 3399 - 3411,2014
26. P.S. Rossi, D. Ciuonzo, T. Ekman, and H. Dong, “Energy Detection for MIMO Decision Fusion in Underwater Sensor Networks,” IEEE Sensors Journal, vol.15, pp.1630-1640, 2014.
27. K.-H. Liu, S. Yan, and C.-C. Kuo, “Age Estimation via Grouping and Decision Fusion” IEEE Transactions on Information Forensics and Security, vol 10, pp 2408-2423, 2015.
28. P. J. Moore, I. E. Portugues, and I. A. Glover, “Radiometric Location of Partial Discharge Sources on Energized High-Voltage Plant,” IEEE Transactions on Power Delivery, vol.33, pp.2047-2049, 2018.